Langenbeck's Archives of Surgery

, Volume 399, Issue 1, pp 127–133

Technical feasibility of liver transplantation without cold storage


    • Department of General, Visceral, and Transplantation Surgery, Charité Campus VirchowCharité Universitätsmedizin Berlin
  • F. Klein
    • Department of General, Visceral, and Transplantation Surgery, Charité Campus VirchowCharité Universitätsmedizin Berlin
  • G. Puhl
    • Department of General, Visceral, and Transplantation Surgery, Charité Campus VirchowCharité Universitätsmedizin Berlin
  • P. Neuhaus
    • Department of General, Visceral, and Transplantation Surgery, Charité Campus VirchowCharité Universitätsmedizin Berlin
Rapid Communication

DOI: 10.1007/s00423-013-1150-x

Cite this article as:
Gül, S., Klein, F., Puhl, G. et al. Langenbecks Arch Surg (2014) 399: 127. doi:10.1007/s00423-013-1150-x



The success of liver transplantation (LT) is accompanied by an increased need for organs. The wider use of older donors and marginal organs with risk factors such as steatosis has lead to a new interest to improve the outcome with marginal organs. We herewith report a novel technique for LT with in situ preparation and immediate warm-ischemia liver transplantation (WI-LT). The aim of our study was to demonstrate the technical feasibility and report the transplant course.


Six patients underwent WI-LT at our institution. Hepatectomies during procurement and LT were both performed in parallel by different surgical teams. Technical factors and postoperative allograft function were analyzed.


All six WI-LTs were performed without intraoperative complications with a mean warm-ischemia time (WIT) of 29.0 min. No patient developed primary non-function or required retransplantation. Mean alanine aminotransferase (194.0 ± 170.4 U/l) and aspartate aminotransferase (316.3 ± 222.1 U/l) values on the first postoperative day were low, indicating a low ischemia/reperfusion injury and an excellent liver function.


These results demonstrate that WI-LT is a safe and technically feasible approach for LT with possibly reduced IRI and an excellent postoperative allograft quality. WI-LT may therefore be considered in individual patients especially with extended criteria donors to eventually improve postoperative allograft quality.


Cold ischemiaIschemia/reperfusion injury (IRI)Liver transplantationOrgan preservationWarm ischemia


Orthotopic liver transplantation (LT) remains the standard treatment for patients with end-stage liver disease and irreversible acute liver failure [1, 2]. Progress in surgical techniques, organ preservation, immunosuppression, and anti-infective therapies has enabled significant improvements in liver transplant outcomes with 1-year patient survival rates of 90 % and 1-year allograft survival rates of 80 % [3]. The increasing success of LT is however confronted with an increasing shortage of available organs. This has lead to a major effort on extending the cohort of possible organ donors by including liver allografts of extended criteria donors (ECD), which exceed the traditional limits for steatosis and/or donor age or are procured via donation after cardiac death (DCD). Because of their sensitivity to preservation-based ischemia/reperfusion injury (IRI), ECD livers are however associated with an increased risk of early allograft dysfunction (EAD) or primary non-function (PNF) after LT [4]. The wider use of marginal organs in recent years has therefore been accompanied by a resurgence of interest in strategies to optimize the condition of available organs during the preservation period. Standard preservation techniques include warm-ischemia time (WIT) and cold ischemia time (CIT) of the allograft and are therefore associated with proportional IRI as a key determinant of postoperative outcome after LT. Besides causing up to 10 % of early transplant failures, IRI is associated with intra- and/or extrahepatic non-anastomotic biliary strictures in the early and long-time postoperative course after LT [57]. Warm IRI occurs during the cessation of liver perfusion and/or during reduced liver perfusion after shock or heart failure, whereas cold IRI is not just proportional to the duration of ischemia but—despite effective preservation solutions—also directly related to the cooling process itself [812]. This is explained by a continuance of anaerobic metabolism along with a depletion of energy stores during the cooling process, which leads to a loss of transcellular electrolyte gradients, an influx of free calcium, a subsequent activation of phospholipases, and therefore to subsequent cell swelling and lysis [5, 11]. Furthermore, cell ischemia mediates the production of toxic molecules after reperfusion, particularly reactive oxygen intermediates, which generate a cascade of events that induce IRI [12]. In an attempt to avoid cold IRI and possibly improve allograft quality and outcome, we developed a novel technique of liver transplantation: after graft procurement without cold preservation, which is based on the setting of an in-house donor and a consecutive in situ preparation of the liver in order to be able to perform an immediate warm-ischemia-only liver transplantation (WI-LT) without cold preservation. The aims of this Rapid Communication are to demonstrate the technical feasibility of this novel approach and to address possible benefits with regard to postoperative allograft function.

Patients and methods


In our study, we analyzed six patients who underwent WI-LT at our institution. The decision to perform WI-LT was based on the availability of six “in-house donors”, which were routinely allocated by Eurotransplant to six patients of our clinic, as well as on biochemical parameters of the patients and on the experience of our transplant surgeons. The degree of liver failure and the priority of the patients were categorized according to the Mayo End-Stage Liver Disease (MELD) score [13]. The six patients who underwent WI-LT were retrospectively analyzed with regard to patient criteria, technical feasibility, ischemia times, intra- and postoperative course, and postoperative laboratory values as indicators for allograft quality and function.

Donor surgical technique

The organ procurement was in each case performed using a modified conventional technique by surgeons who are experienced in LT as well as in extended liver resection. The donor hemodynamic system was closely monitored to especially maintain a constant kidney and heart perfusion. After opening of the abdominal cavity by median laparotomy, the focus was directed on specific and accurate in situ liver preparation. This included the complete dissection and circulation of the supra- and infrahepatic vena cava (SVC, IVC), the hepatic artery (HA), the portal vein (PV), and the common bile duct (BD). We started by carefully dissecting the celiac trunk free. Herefore, the first 2 cm of the splenic artery was exposed, and the common hepatic artery was then entirely dissected. Afterwards, the gastroduodenal artery was divided between ligatures. In a next step, the portal vein was freed up from the confluens region to the portal vein bifurcation. Then, the common bile duct was cut above the pancreas without any ligature, and the right adrenal vein and any other vessels from the retroperitoneum to the vena cava were divided between ligatures. After systemic intravenous application of 25,000 U of heparin, the portal vein was clamped above the pancreas and divided and the celiac trunk and the splenic artery were clamped. The common hepatic artery was then intersected at its origin at the celiac trunk and cut in a bell shape as a preparation for the later arterial anastomosis. Finally, the vena cava was clamped infrahepatically and suprahepatically, and then, the liver was excised without any kind of perfusion so that the warm soft organ, filled with heparinized blood, could be brought to the operation room next door where the recipient hepatectomy had already been completed. After the liver was removed, the standard intra-aortal perfusion solution sequence for the kidneys was subsequently started in accordance to international standards [14].

Recipient surgical technique

The recipient operation was begun in parallel to the donor operation with frequent communication between the operating teams. Both hepatectomies were performed in parallel to be able to proceed with immediate liver transplantation after warm-ischemia graft procurement and preparation and thereby keep the liver warm-ischemia time as short as possible. LT was performed using a standardized technique with inferior vena cava interposition in five patients and by piggyback technique in one patient. The vessels in the donor were clamped and divided after completion of the hepatectomy and achievement of complete hemostasis in the recipient. Reperfusion was started after the portal vein was anastomosed in an end-to-end fashion. Subsequently, venting was started via the inferior cava vein. The arterial and the remaining caval anastomoses were performed after hemostasis had been achieved. Side-to-side common bile duct anastomosis was performed, and a t-tube was placed in the common bile duct of the recipient.

Ischemia times

Warm-ischemia time was documented by the operating surgeon and per definition started with cross-clamping during procurement in the donor and ended with portal venous reperfusion in the recipient.

Standard postoperative care

All patients were referred to an intensive care unit (ICU) postoperatively and treated by standard LT protocol. Immunosuppression was started with FK506 and prednisolone. Infectious and antiviral prophylaxis was administered systemically. Color Doppler sonography was performed daily on the first five postoperative days. X-ray cholangiography was performed routinely 5 days and 6 weeks after LT via the t-tube. Liver biopsies were routinely performed in all six patients 6 months after LT.

For our analysis, a formal consent was obtained from the patients and the chief of the surgical department. Eurotransplant was informed about the protocol and the selection of the patients.



The median age of the patients who underwent WI-LT was 52.4 years, and two patients were male. The indications for LT were alcoholic cirrhosis in one patient (16 %), hepatitis B cirrhosis in one patient (16 %), and hepatitis C cirrhosis in four patients (67 %). The median donor age was 48.2 years, and two of the six donors were male (33 %).

Ischemia times and intraoperative findings

The mean WIT was 29.0 min with no CIT. The mean operation time was 268.5 min (230–319) (Table 1) with a mean transfusion amount of 5.2 erythrocyte concentrates and 14.3 fresh frozen plasma concentrates. No intraoperative complications were observed.
Table 1

Operative characteristics and important posttransplant findings

Mean operation time (range)

268.5 min (230–319)

WIT min (median, SD, range)

29.0 ± 8.0 (18–41)

Major morbidity

Vascular thrombosis



Hospital stay (days) (SD)

31.2 ± 16.0 (16–60)

Biliary complications (stenosis)





Overall mortality (n)

Postoperative laboratory findings

Mean alanine aminotransferase (ALT; U/l) level was 194.0 (±170.4) on the first postoperative day and 90.8 (±100.2) on the sixth postoperative day (Fig. 1), whereas mean aspartate aminotransferase (AST; U/l) levels were 316.3 (±222.1) and 68.8 (±131.4), respectively (Fig. 1). Levels of total serum bilirubin (mg/dl) and PT (%) were 4.4 (±3.2) and 71.8 (±5.0) on the first postoperative day versus 1.7 (±1.2) and 81.8 (±14.2) on the sixth postoperative day (Figs. 2 and 3).
Fig. 1

Median laboratory values for ALT and AST (postoperative days 1 to 7)
Fig. 2

Median laboratory values for bilirubin (postoperative days 1 to 7)
Fig. 3

Median laboratory values for PT (postoperative days 1 to 7)

Postoperative clinical course

All patients had excellent postoperative bile production (Fig. 4). No major morbidities according to the Clavien classification were observed. The mean length of stay in the intensive care unit was 7.0 days (±3.4), and the mean overall length of hospital stay was 29.5 (±5.7) days. One patient (16.7 %) underwent postoperative endoscopic cholangiography (ERC) with endoscopic stent placement because of a relevant common bile duct stenosis in the postoperative course. No ischemic type biliary lesions (ITBL) were observed. Grade 1 rejection according to the Banff scheme was observed in two patients (33.3 %) and successfully treated with methylprednisolone (500 mg) for 3 days. A liver biopsy within 6 months after LT was performed in all six patients of the WI group and revealed fibrosis grade > II in four patients (66.7 %) and hepatic steatosis in four patients (66.7 %) (Table 1).
Fig. 4

Median postoperative bile production (postoperative days 1 to 5). The t-tube was closed on the fifth postoperative day


Orthotopic liver transplantation remains the most effective therapy for end-stage liver disease but is limited by the persistent shortage of suitable donor organs. This shortage has prompted the increased use of ECD livers, which are associated with an increased risk of EAD or PNF after LT due to severe IRI [4]. Improvements in the effectiveness of organ preservation are therefore essential in order to expand the cohort of possible organ donors. Cold storage (CS) at 4 °C is the current standard method of preserving liver grafts for LT and has been unchanged for two decades. The preservation solutions used for CS typically include colloids, antioxidants, and molecular precursors of ATP that help to reduce the environmental pressures imposed by hypothermia. However, even with the most effective preservation solution, CS leads to immediate graft injury at the time of transplantation. This injury not just occurs proportionally to the duration of ischemia but is also specifically related to the cooling process [1517]. This is underlined by the results of Olschewski et al. who investigated the influence of perfusate temperature during oxygenated machine perfusion on the graft quality and found that an increase in the perfusion temperature from 4 °C to 21 °C resulted in a marked reduction of portal venous resistance and an increased bile production [18]. Reddy et al. found that even 1 h of cold preservation already lead to a significant graft injury by hepatocellular damage, Kupffer cell activation, and sinusoidal endothelial cell dysfunction [19]. A high degree of liver steatosis may be a negative key determinant in this context, especially in extended criteria donors, because steatotic livers exhibit exacerbated endoplasmic reticulum stress, which occurs in response to cold IRI. This is leading to a defective liver autophagy and therefore consecutive liver damage [20]. Furthermore, the differences in fluidity between fat drops and the cytoplasm during the cooling process and in low temperature may lead to shear stress of the tissue compound and therefore aggravate allograft damage especially in steatotic livers.

The awareness of these adverse effects of cold preservation has lead to renaissance of normothermic machine perfusion with diluted, hyperbaric oxygenated blood as performed by Starzl during the first successful human liver transplantation [21]. In contrast to the concept of CS preservation, the concept of normothermic preservation is not only to maintain cellular metabolism, but also to avoid direct damage by tissue cooling. There is accumulating evidence of the superiority of this more physiological approach in association with an oxygenated blood-based perfusion solution. However, in clinical practice, machine perfusion is until today only carried out in the field of kidney transplantation, in particular for extended-criteria kidney grafts [22, 23]. The technique of WI-LT avoids CS damage and allowed us to achieve a median WIT of 29 min. This approach seems convenient on one hand and logistically challenging on the other hand. The strategy was based on using two parallel surgical teams during LT so that the recipient team could commence surgery as soon as the donor team confirmed that the liver was “useable”. From a surgical technical perspective, the in situ preparation and liver procurement during WI-LT are performed analogously to standard dissection techniques in liver surgery. Pringle maneuver-related total liver ischemia times of up to 120 min have been reported as safe in this context [24]. The median WIT of 29 min for WI-LT in our study may therefore be seen as well tolerable.

The patients selected for this type of LT were carefully evaluated. A difficult hepatectomy, hepatic portal vein thrombosis, and hepatic arterial problems were considered contraindications for this procedure because with increasing surgical difficulty, the WIT may exceed limits of IRI. Despite the logistical challenge of this novel approach, the results of our study are seen as promising. All recipients had an uneventful intra- and postoperative course. An excellent postoperative bile production and, in addition, early laboratory findings for PT, ALT and AST, and bilirubin also indicated an excellent graft function underlining a possible low degree of IRI. No PNF occurred and no patient required retransplantation. One patient required endoscopic bile duct stent placement due to a relevant bile duct stenosis, but no patient developed ITBL, which usually presents with symptoms within 6 months after LT and therefore corresponds to the observation period of our study [25, 26]. In a time in which we are challenged with an increasing demand for organs to treat end-stage-liver disease, we need to extend our pool of possible organs, not just by the extension of the donation criteria but also by improving our preservation techniques to minimize IRI in available organs, especially in ECD organs. As a result of our study, WI-LT was shown to be technically feasible and safe, with a positive impact on allograft quality and postoperative outcome. In the setting of an in-house donor and in-house recipient organ allocation, WI-LT may therefore be considered as a technique in individually selected patients to further improve allograft quality and postoperative outcomes after LT, especially in patients with ECD. Due to the logistical and technical challenges of this approach, WI-LT should however only be performed at experienced liver transplant centers. This is of course only a pilot study to investigate the technical feasibility and early postoperative outcome of this novel technique. Further prospective clinical trials are therefore necessary to justify a routine clinical application of this novel technique in order to possibly establish a more broad clinical application of WI-LT.

Conflicts of interest

The authors of this Rapid Communication have no conflicts of interest to disclose. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

Ethical approval

Our study was fully approved by the ethical commission of our institution.

Copyright information

© Springer-Verlag Berlin Heidelberg 2013